Group modeling for fmri
Download
1 / 56

Group Modeling for fMRI - PowerPoint PPT Presentation


  • 185 Views
  • Uploaded on

Group Modeling for fMRI. Thomas Nichols, Ph.D. Assistant Professor Department of Biostatistics http://www.sph.umich.edu/~nichols IPAM MBI July 21, 2004. Overview. Mixed effects motivation Evaluating mixed effects methods Case Studies Summary statistic approach (HF) SPM2 FSL3

loader
I am the owner, or an agent authorized to act on behalf of the owner, of the copyrighted work described.
capcha
Download Presentation

PowerPoint Slideshow about 'Group Modeling for fMRI' - salena


An Image/Link below is provided (as is) to download presentation

Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author.While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server.


- - - - - - - - - - - - - - - - - - - - - - - - - - E N D - - - - - - - - - - - - - - - - - - - - - - - - - -
Presentation Transcript
Group modeling for fmri
Group Modelingfor fMRI

Thomas Nichols, Ph.D.

Assistant Professor

Department of Biostatistics

http://www.sph.umich.edu/~nichols

IPAM MBI

July 21, 2004


Overview
Overview

  • Mixed effects motivation

  • Evaluating mixed effects methods

  • Case Studies

    • Summary statistic approach (HF)

    • SPM2

    • FSL3

  • Conclusions


Overview1
Overview

  • Mixed effects motivation

  • Evaluating mixed effects methods

  • Case Studies

    • Summary statistic approach (HF)

    • SnPM

    • SPM2

    • FSL3

  • Conclusions


Lexicon
Lexicon

Hierarchical Models

  • Mixed Effects Models

  • Random Effects (RFX) Models

  • Components of Variance

    ... all the same

    ... all alluding to multiple sources of variation

    (in contrast to fixed effects)


Random effects illustration

Standard linear modelassumes only one source of iid random variation

Consider this RT data

Here, two sources

Within subject var.

Between subject var.

Causes dependence in 

3 Ss, 5 replicated RT’s

Random Effects Illustration

Residuals


Fixed vs random effects in fmri

Distribution of each subject’s estimated effect

Fixed vs.RandomEffects in fMRI

2FFX

Subj. 1

Subj. 2

  • Fixed Effects

    • Intra-subject variation suggests all these subjects different from zero

  • Random Effects

    • Intersubject variation suggests population not very different from zero

Subj. 3

Subj. 4

Subj. 5

Subj. 6

0

2RFX

Distribution of population effect


Fixed effects
Fixed Effects

  • Only variation (over subjects/sessions) is measurement error

  • True Response magnitude is fixed


Random mixed effects
Random/Mixed Effects

  • Two sources of variation

    • Measurement error

    • Response magnitude

  • Response magnitude is random

    • Each subject/session has random magnitude


Random mixed effects1
Random/Mixed Effects

  • Two sources of variation

    • Measurement error

    • Response magnitude

  • Response magnitude is random

    • Each subject/session has random magnitude

    • But note, population mean magnitude is fixed


Fixed vs random
Fixed vs. Random

  • Fixed isn’t “wrong,” just usually isn’t of interest

  • Fixed Effects Inference

    • “I can see this effect in this cohort”

  • Random Effects Inference

    • “If I were to sample a new cohort from the population I would get the same result”


Overview2
Overview

  • Mixed effects motivation

  • Evaluating mixed effects methods

  • Case Studies

    • Summary statistic approach (HF)

    • SnPM

    • SPM2

    • FSL3

  • Conclusions


Assessing rfx models issues to consider
Assessing RFX ModelsIssues to Consider

  • Assumptions & Limitations

    • What must I assume?

    • When can I use it

  • Efficiency & Power

    • How sensitive is it?

  • Validity & Robustness

    • Can I trust the P-values?

    • Are the standard errors correct?

    • If assumptions off, things still OK?


Issues assumptions
Issues: Assumptions

  • Distributional Assumptions

    • Gaussian? Nonparametric?

  • Homogeneous Variance

    • Over subjects?

    • Over conditions?

  • Independence

    • Across subjects?

    • Across conditions/repeated measures

    • Note:

      • Nonsphericity = (Heterogeneous Var) or (Dependence)


Issues soft assumptions regularization
Issues: Soft AssumptionsRegularization

  • Regularization

    • Weakened homogeneity assumption

    • Usually variance/autocorrelation regularized over space

  • Examples

    • fmristat - local pooling (smoothing) of (2RFX)/(2FFX)

    • SnPM - local pooling (smoothing) of 2RFX

    • FSL3 - Bayesian (noninformative) prior on 2RFX


Issues efficiency power
Issues: Efficiency & Power

  • Efficiency: 1/(Estmator Variance)

    • Goes up with n

  • Power: Chance of detecting effect

    • Goes up with n

    • Also goes up with degrees of freedom (DF)

      • DF accounts for uncertainty in estimate of 2RFX

      • Usually DF and n yoked, e.g. DF = n-p


Issues validity
Issues: Validity

  • Are P-values accurate?

    • I reject my null when P < 0.05Is my risk of false positives controlled at 5%?

    • “Exact” control

      • FPR = a

    • Valid control (possibly conservative)

      • FPR  a

  • Problems when

    • Standard Errors inaccurate

    • Degrees of freedom inaccurate


Overview3
Overview

  • Mixed effects motivation

  • Evaluating mixed effects methods

  • Case Studies

    • Summary statistic approach (HF)

    • SnPM

    • SPM2

    • FSL3

  • Conclusions


Four rfx approaches in fmri
Four RFX Approaches in fMRI

  • Holmes & Friston (HF)

    • Summary Statistic approach (contrasts only)

    • Holmes & Friston (HBM 1998). Generalisability, Random Effects & Population Inference. NI, 7(4 (2/3)):S754, 1999.

  • Holmes et al. (SnPM)

    • Permutation inference on summary statistics

    • Nichols & Holmes (2001). Nonparametric Permutation Tests for Functional Neuroimaging: A Primer with Examples. HBM, 15;1-25.

    • Holmes, Blair, Watson & Ford (1996). Nonparametric Analysis of Statistic Images from Functional Mapping Experiments. JCBFM, 16:7-22.

  • Friston et al. (SPM2)

    • Empirical Bayesian approach

    • Friston et al. Classical and Bayesian inference in neuroimagining: theory. NI 16(2):465-483, 2002

    • Friston et al. Classical and Bayesian inference in neuroimaging: variance component estimation in fMRI. NI: 16(2):484-512, 2002.

  • Beckmann et al. & Woolrich et al. (FSL3)

    • Summary Statistics (contrast estimates and variance)

    • Beckmann, Jenkinson & Smith. General Multilevel linear modeling for group analysis in fMRI. NI 20(2):1052-1063 (2003)

    • Woolrich, Behrens et al. Multilevel linear modeling for fMRI group analysis using Bayesian inference. NI 21:1732-1747 (2004)


Overview4
Overview

  • Mixed effects motivation

  • Evaluating mixed effects methods

  • Case Studies

    • Summary statistic approach (HF)

    • SnPM

    • SPM2

    • FSL3

  • Conclusions


Case studies holmes friston
Case Studies:Holmes & Friston

  • Unweighted summary statistic approach

  • 1- or 2-sample t test on contrast images

    • Intrasubject variance images not used (c.f. FSL)

  • Proceedure

    • Fit GLM for each subject i

    • Compute cbi, contrast estimate

    • Analyze {cbi}i


Case studies hf assumptions
Case Studies: HFAssumptions

  • Distribution

    • Normality

    • Independent subjects

  • Homogeneous Variance

    • Intrasubject variance homogeneous

      • 2FFX same for all subjects

    • Balanced designs


Case studies hf limitations

From HBMPosterWE 253

Case Studies: HFLimitations

  • Limitations

    • Only single image per subject

    • If 2 or more conditions,Must run separate model for each contrast

  • Limitation a strength!

    • No sphericity assumption made on conditions

    • Though nonsphericity itself may be of interest...


Case studies hf efficiency
Case Studies: HFEfficiency

  • If assumptions true

    • Optimal, fully efficient

  • If 2FFX differs between subjects

    • Reduced efficiency

    • Here, optimal requires down-weighting the 3 highly variable subjects

0


Case studies hf validity
Case Studies: HFValidity

  • If assumptions true

    • Exact P-values

  • If 2FFX differs btw subj.

    • Standard errors OK

      • Est. of 2RFX unbiased

    • DF not OK

      • Here, 3 Ss dominate

      • DF < 5 = 6-1

0

2RFX


Case studies hf robustness
Case Studies: HFRobustness

  • Heterogeneity of 2FFX across subjects...

    • How bad is bad?

  • Dramatic imbalance (rough rules of thumb only!)

    • Some subjects missing 1/2 or more sessions

    • Measured covariates of interest having dramatically different efficiency

      • E.g. Split event related predictor by correct/incorrect

      • One subj 5% trials correct, other subj 80% trials correct

  • Dramatic heteroscedasticity

    • A “bad” subject, e.g. head movement, spike artifacts


Overview5
Overview

  • Mixed effects motivation

  • Evaluating mixed effects methods

  • Case Studies

    • Summary statistic approach (HF)

    • SnPM

    • SPM2

    • FSL3

  • Conclusions


Case studies snpm

5%

Parametric Null Distribution

5%

Nonparametric Null Distribution

Case Studies: SnPM

  • No Gaussian assumption

  • Instead, uses data to find empirical distribution


Permutation test toy example
Permutation TestToy Example

  • Data from V1 voxel in visual stim. experiment

    A: Active, flashing checkerboard B: Baseline, fixation

    6 blocks, ABABAB Just consider block averages...

  • Null hypothesis Ho

    • No experimental effect, A & B labels arbitrary

  • Statistic

    • Mean difference


Permutation test toy example1
Permutation TestToy Example

  • Under Ho

    • Consider all equivalent relabelings


Permutation test toy example2
Permutation TestToy Example

  • Under Ho

    • Consider all equivalent relabelings

    • Compute all possible statistic values


Permutation test toy example3
Permutation TestToy Example

  • Under Ho

    • Consider all equivalent relabelings

    • Compute all possible statistic values

    • Find 95%ile of permutation distribution


Permutation test toy example4
Permutation TestToy Example

  • Under Ho

    • Consider all equivalent relabelings

    • Compute all possible statistic values

    • Find 95%ile of permutation distribution


Permutation test toy example5
Permutation TestToy Example

  • Under Ho

    • Consider all equivalent relabelings

    • Compute all possible statistic values

    • Find 95%ile of permutation distribution

-8

-4

0

4

8


Case studies snpm assumptions
Case Studies: SnPMAssumptions

  • 1-Sample t on difference data

    • Under null, differences distn symmetric about 0

      • OK if 2FFX differs btw subjects!

    • Subjects independent

  • 2-Sample t

    • Under null, distn of all data the same

      • Implies 2FFX must be the same across subjects

    • Subjects independent


Case studies snpm efficiency
Case Studies: SnPMEfficiency

  • Just as with HF...

    • Lower efficiency if heterogeneous variance

    • Efficiency increases with n

    • Efficiency increases with DF

      • How to increase DF w/out increasing n?

      • Variance smoothing!


Case studies snpm smoothed variance t

mean difference

Case Studies: SnPMSmoothed Variance t

  • For low df, variance estimates poor

    • Variance image “spikey”

  • Consider smoothed variance t statistic

t-statistic

variance


Permutation test smoothed variance t
Permutation TestSmoothed Variance t

  • Smoothed variance estimate better

    • Variance “regularized”

    • df effectively increased, but no RFT result

SmoothedVariancet-statistic

mean difference

smoothedvariance


Small group fmri example

12 subject study

Item recognition task

1-sample t test on diff. imgs

FWE threshold

uRF = 9.87uBonf = 9.805 sig. vox.

uPerm = 7.67 58 sig. vox.

378 sig. vox.

Smoothed Variance t Statistic,Nonparametric Threshold

t11Statistic, Nonparametric Threshold

t11Statistic, RF & Bonf. Threshold

Permutation Distn Maximum t11

Small GroupfMRI Example

Working memory data: Marshuetz et al (2000)


Case studies snpm validity
Case Studies: SnPMValidity

  • If assumptions true

    • P-values’s exact

  • Note on “Scope of inference”

    • SnPM can make inference on population

    • Simply need to assume random sampling of population

      • Just as with parametric methods!


Case studies snpm robustness
Case Studies: SnPMRobustness

  • If data not exchangeable under null

    • Can be invalid, P-values too liberal

    • More typically, valid but conservative


Overview6
Overview

  • Mixed effects motivation

  • Evaluating mixed effects methods

  • Case Studies

    • Summary statistic approach (HF)

    • SnPM

    • SPM2

    • FSL3

  • Conclusions


Case study spm2
Case Study: SPM2

  • 1 effect per subject

    • Uses Holmes & Friston approach

  • >1 effect per subject

    • Variance basis function approach used...


Spm2 notation iid case

12 subjects,4 conditions

Use F-test to find differences btw conditions

Standard Assumptions

Identical distn

Independence

“Sphericity”... but here not realistic!

SPM2 Notation: iid case

y = X + e

N 1 N  pp  1 N  1

X

Error covariance

N

N


Multiple variance components

12 subjects, 4 conditions

Measurements btw subjects uncorrelated

Measurements w/in subjects correlated

Multiple Variance Components

y = X + e

N 1 N  pp  1 N  1

Error covariance

N

N

Errors can now have

different variances and

there can be correlations

Allows for ‘nonsphericity’


Non sphericity modeling

Errors are independent

but not identical

Errors are not independent

and not identical

Non-Sphericity Modeling

Error Covariance


Case study spm21
Case Study: SPM2

  • Assumptions & Limitations

    • assumed to globallyhomogeneous

    • lk’s only estimated from voxels with large F

    • Most realistically, Cor(e) spatially heterogeneous

    • Intrasubject variance assumed homogeneous


Case study spm22
Case Study: SPM2

  • Efficiency & Power

    • If assumptions true, fully efficient

  • Validity & Robustness

    • P-values could be wrong (over or under) if local Cor(e) very different from globally assumed

    • Stronger assumptions than Holmes & Friston


Overview7
Overview

  • Mixed effects motivation

  • Evaluating mixed effects methods

  • Case Studies

    • Summary statistic approach (HF)

    • SnPM

    • SPM2

    • FSL3

  • Conclusions


Fsl3 full mixed effects model
FSL3: Full Mixed Effects Model

First-level, combines sessions

Second-level, combines subjects

Third-level, combines/compares groups



Summary stats equivalence

Crucially, summary stats here are not just estimated effects.Summary Stats needed for equivalence:

Summary Stats Equivalence

Beckman et al., 2003


Case study fsl3 s flame
Case Study: FSL3’s FLAME

  • Uses summary-stats model equivalent to full Mixed Effects model

  • Doesn’t assume intrasubject variance is homogeneous

    • Designs can be unbalanced

    • Subjects measurement error can vary


Case study fsl3 s flame1
Case Study: FSL3’s FLAME

  • Bayesian Estimation

    • Priors, priors, priors

    • Use reference prior

  • Final inference on posterior of b

    • b | y has Multivariate T distn (MVT)but with unknown dof


Approximating mvts

Gaussian

dof?

BIDET

MCMC

Approximating MVTs

FAST

Estimate

MVT

Model

Samples

BIDET = Bayesian Inference with Distribution Estimation using T

SLOW


Overview8
Overview

  • Mixed effects motivation

  • Evaluating mixed effects methods

  • Case Studies

    • Summary statistic approach (HF)

    • SnPM

    • SPM2

    • FSL3

  • Conclusions


Conclusions
Conclusions

  • Random Effects crucial for pop. inference

  • Don’t fall in love with your model!

  • Evaluate...

    • Assumptions

    • Efficiency/Power

    • Validity & Robustness


ad